Edge computing critical to success of autonomous vehicles
Edge computing will play a compelling role in autonomous vehicles becoming mainstream in the near future.
This is according to Kevin Brown, senior VP of innovation and secure power division and CTO at multinational energy management and automation solutions provider, Schneider Electric.
Speaking this week at the “Life at the Edge” media conference in Singapore, Brown discussed the role of edge computing in self-driving or autonomous vehicles.
He explained that while there is a long way to go before fully autonomous vehicles become mainstream across the globe, edge computing plays an important role in enabling its main capabilities. These include on-board computing tasks, which use terabytes of data to enable the vehicle to communicate with numerous networks and devices around it, allowing it to be constantly aware of its environment and make split-second decisions.
“The automotive vertical presents some of the most exciting edge computing use cases on how edge computing optimises smart devices such as sensors to store and process information. We are seeing the adoption of innovative technologies in every phase of the automotive sector, from the manufacturing stage – while it’s still in the factory – all the way to the distribution process, and ultimately to the customer experience stage, during purchasing.”
The vision of the autonomous vehicle becoming mainstream can only be realised when a vehicle has the ability to anticipate what’s happening around it and predict what is going to happen. This is made possible by a combination of computing power, a low latency network and sensors, to provide situational awareness, explained Brown.
“Autonomous vehicles provide a great edge application, because data storage and data processing take place closer to the source of data and not at a large centralised system.
“While the vehicle is trying to process what’s happening around it using this data, many sensors are used to understand what’s going on, what’s at risk, and what decisions should be made.
“That’s why when an accident occurs, it’s usually because those sensors don’t recognise what’s happening. Therefore a crucial enabling element is the ‘micro data centres’ purely built into the car. However, other supporting infrastructure around that, such as 5G, will determine its performance.”
In recent years, there has been an incredible shift in the number of software developers found in a vehicle manufacturing company, versus the number of mechanical engineers one can find in the same company, indicating an increase in the importance of software developers in the industry, he continued.
Jim Simonelli, SVP of emerging businesses: secure power division at Schneider Electric, explained the future of autonomous vehicles will take place in three waves.
“The first stage isautomated, which is what is going on right now in both current fleets and in testing phases, where many of these pilots involve edge computing.
“The second stage is autonomous, where a driver assistance system is used. The third stage is completely driverless, where the car is capable of taking full control without human intervention.”
The Singapore Grand Prix, which forms part of the FIA Formula One World Championship, taking place this weekend, uses edge computing technology to improve performance in all its vehicles, noted Simonelli.
“Millions of dollars are spent on the F1 race vehicles to give them a competitive advantage, such as storing up energy, keeping track of the driver’s physical wellbeing through sensors, enhancing performance and keeping track of the condition of the vehicle.”